chaoqin-li1123 commented on code in PR #45023:
URL: https://github.com/apache/spark/pull/45023#discussion_r1517356072
##########
python/pyspark/sql/datasource.py:
##########
@@ -298,6 +320,133 @@ def read(self, partition: InputPartition) ->
Iterator[Union[Tuple, Row]]:
...
+class DataSourceStreamReader(ABC):
+ """
+ A base class for streaming data source readers. Data source stream readers
are responsible
+ for outputting data from a streaming data source.
+
+ .. versionadded: 4.0.0
+ """
+
+ def initialOffset(self) -> dict:
+ """
+ Return the initial offset of the streaming data source.
+ A new streaming query starts reading data from the initial offset.
+ If Spark is restarting an existing query, it will restart from the
check-pointed offset
+ rather than the initial one.
+
+ Returns
+ -------
+ dict
+ A dict or recursive dict whose key and value are primitive types,
which includes
+ Integer, String and Boolean.
+
+ Examples
+ --------
+ >>> def initialOffset(self):
+ ... return {"parititon-1": {"index": 3, "closed": True},
"partition-2": {"index": 5}}
+ """
+
+ ...
+ raise PySparkNotImplementedError(
+ error_class="NOT_IMPLEMENTED",
+ message_parameters={"feature": "initialOffset"},
+ )
+
+ def latestOffset(self) -> dict:
+ """
+ Returns the most recent offset available.
+
+ Returns
+ -------
+ dict
+ A dict or recursive dict whose key and value are primitive types,
which includes
+ Integer, String and Boolean.
+
+ Examples
+ --------
+ >>> def latestOffset(self):
+ ... return {"parititon-1": {"index": 3, "closed": True},
"partition-2": {"index": 5}}
+ """
+ ...
+ raise PySparkNotImplementedError(
+ error_class="NOT_IMPLEMENTED",
+ message_parameters={"feature": "latestOffset"},
+ )
+
+ def partitions(self, start: dict, end: dict) -> Sequence[InputPartition]:
+ """
+ Returns a list of InputPartition given the start and end offsets.
Each InputPartition
+ represents a data split that can be processed by one Spark task.
+
+ Parameters
+ ----------
+ start : dict
+ The start offset of the microbatch to plan partitioning.
+ end : dict
+ The end offset of the microbatch to plan partitioning.
+
+ Returns
+ -------
+ Sequence[InputPartition]
+ A sequence of partitions for this data source. Each partition value
+ must be an instance of `InputPartition` or a subclass of it.
+ """
+ ...
+ raise PySparkNotImplementedError(
+ error_class="NOT_IMPLEMENTED",
+ message_parameters={"feature": "partitions"},
+ )
+
+ @abstractmethod
+ def read(self, partition: InputPartition) -> Iterator[Union[Tuple, Row]]:
Review Comment:
I copy the interface from batch python data source. Why is this
Iterator[Union] instead of Union[Iterator]? @HyukjinKwon
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
